Translation≠Localization

This is a quick heads-up about a common misconception: The idea that you can simply translate your website content into another language using software.

There are a number of powerful machine translation tools out there, such as DeepL, Google Translate, or even ChatGPT. Many of them can even be embedded on your site to automatically translate your content — theoretically allowing you to offer your site in many of then languages your audience might speak.

But before you do so, you should understand that translation isn’t the same as localization — and their difference has significant implications for your website’s performance and your brand.

Strictly speaking, translation means converting words from one language into another. In practice, modern software is able to do this relatively well—rather than just robotically changing each word, it also rearranges the sentence structures and tweaks the grammar for the translated sentence to read correctly.

Machine translation is great for quickly grasping the meaning or context of a text. And yet, automatic translations usually have a few problems. As direct translations of your content, they often won’t read like authentic, native-sounding text. They often don’t have a natural flow, meaning they’ll feel obviously translated. This is particularly true for the more idiosyncratic parts of a text, such as sayings and figures of speech, or cultural references that likely don’t make sense in another language.

This brings us to the topic of localization. A localized text isn’t just translated, but it has been changed or rewritten to make it local in character. It conveys the same message, but reads as if it were originally written in the target language. It means that it will use different figures of speech, different metaphors and different references to convey that information.

While machine translation has become very accurate in conveying information, it (still) struggles with true localization. Since software doesn’t have a native speaker’s cultural references and intuition for tone and phrasing, it cannot accurately transform the character of a text.

We understand the temptation to use software for translations, particularly for multi-language websites: It’ll save you the process of localizing your content and means you can launch your website quicker. Dynamic, real-time translations even mean that your changes will be reflected in a different language right as you make them.

From a branding perspective, we nevertheless advise against it. One of the core tenets of a strong brand is reliability and trustworthiness: If you are charging for your product or services; if you offer advice or important information, then you simply cannot risk inaccurate translations or improperly localized content. It will affect how your clients perceive your brand — and may damage their trust in it. This, in turn, affects how they interact with your brand or website, and if they’re willing to trust you or pay for your services.

Translation tools powered by generative AI can even produce so-called “hallucinations”, meaning the software will add information in the translation that doesn’t exist in the original.

To protect your brand integrity, it’s essential that your content is properly localized. For now, that still means relying on humans to localize your content — and carefully reviewing everything you publish. That doesn’t mean having to start from scratch with each text: You can use a machine translation as a starting point and then review it with a native speaker to localize its contents and ensure it reads well.

(You better believe we localized this text in the German version.)

Sometimes less is also more: Focus on doing fewer languages well and offer high-quality content based on what your audience actually speaks.

Want content that resonates across cultures — not just languages? Let’s localize your message and make it truly connect. Get in touch.